Well it seems to me that your suggesting a rare occurrence should take precedence over a more common one.
That's what you said before, but I would not say that's what I'm doing. On the contrary, as you are probably aware, I very frequently say that TN-C-S 'lost neutrals' are such a rare occurrence that I don't really understand why people seem to get so excited about them - so I most certainly don't prioritise that very rare risk over anything else!
If there is no earth then how do the RCD's work? do they not need to reference to earth or is it just another wire as far as the RCD is concerned?
Yes, they need 'an earth' to work - but you were talking about (as you do again below) a 'good earth', and I was pointing out (as I had to the OP) that such is really not necessary. For a 30 mA RCD operate only requires an EFLI less than about 7,667 Ω, over 76 times higher than what (100 Ω) would normally be regarded as the minimum acceptable for an earth rod. In other words, and extremely 'bad' earth would be more than adequate to facilitate RCD operation.
And if this is the case then why the fekking heckings have all the electricians etc. been going on and on about having a good earth for so long I cannot remember?
Well, I suppose the simple answer is that you would have to ask them, but I think your problem is that you are probably muddling different contexts....
In
any TN installation with an intact DNO-supplied earth, fault protection via ADS (i.e. provided by OPDs such as MCBs or fuses etc) is totally reliant on that DNO-supplied earth, plus the R1+R2 of the final circuits concerned, resulted in a Zs (from the location of the fault) low enough to result in operation of the OPD (MCB, fuse or whatever) within the required disconnection time, and that requires that the Ze of the DNO-supplied earth be very low (usually saiod to be <0.35 for TN-C-S and <0.8 for TN-S). So, for fault protection by OPD, you certainly need a 'very good' (very low impedance/Ze) DNO-supplied earth, and that's why testing an installation (as, for example, for an EICR) majors on measuring/determining Zs for each circuit, and making sure that it is 'very low' (low enough for the OPD protecting the circuit)..
However, if one 'loses' that DNO earth (either due to a break in a TN-S sheath, or 'loss of the neutral' (CNE) of a TN-C-S supply, then OPD-mediated fault protection cannot happen, so if one does not have RCD protraction, then one will have no fault protection at all. However, as above, if one
does have RCDs, then a 'very bad' earth (e.g. a very poor earth rod) is more than adequate to provide that RCD-mediated protection. In fact, in any installation which has bonded extraneous-c-ps, the path to earth via the extraneous-c-p will be more than enough to facilitate RCD function, even if one has no 'earth rod' at all.
BY the way, if you think that all circuits are RCD protected then your living in cloud cookoo land.
No, I obviously don't think that. However, as above, if one doesn't have RCDs and loses one's DNO-supplied earth, then one inevitably has no fault protection (protection against L-E or L-CPC faults) at all, and nothing will change that - so it's not a situation one can say or do anything about (other than install some RCDs!). In particular, any (domestically) credible 'good earth' will not help - indeed, all a (credibly) 'better' (lower impedance) earth would achieve would be to somewhat
increase the severity (hence increase the risk of death) of any electric shock a person suffered by touching it (and something 'live').
In fact if one doesn't have RCDs and loses one's DNO-supplied earth, then the safest situation of all would be to have to have no 'earth' (good or bad) in the house at all - although that will be impossible if there are bonded extraneous-c-ps.
Is that a bit clearer?
Kind Regards, John
